Previous research indicates that difficulty in sequencing short-duration speech elements is associated with many forms of language-based learning disability (LBLD). In conjunction with psychophysical studies of auditory processing in normal and LBLD subjects, the proposed research will explore the underlying cortical basis of the temporal-processing deficits found in LBLD. The broad goals of these studies are to examine how the brain represents temporal information about auditory stimuli, and to what extent learning can modify such representations. The specific aims are to train monkeys to identify rapid sequences of sound- a specific, ordered series of tones, or a specific tone masked by following noise- followed by recording in primary auditory cortex of neuronal responses to trained and untrained sequences. Frequency- response maps, frequency-tuning curves for individual neurons, and coherence of multi-unit responses will all be examined to answer questions as: Will more neurons respond specifically to each individual element of a trained sequence of stimuli? to the complete ordered sequence? Will the neuronal populations responding to the separate elements of the trained sequence increase the coherence of their responses? Will training the same auditory distinction via two different strategies produce the same or different changes in cortical organization? The proposed experiments will provide considerable information relevant to current issues in both basic and applied research with a design that is economical in its use of time, resources, and animals.